PROJECT SUMMARY Multiple myeloma (MM) is a heterogeneous group of disorders with distinct bone marrow dependence, clinical features, prognosis, and response to therapy. Despite the best available treatments, MM is generally incurable and patients will eventually develop relapsed/refractory disease. Early relapse is associated with poor clinical outcomes and overall survival. There are currently no effective markers that can predict which patients will have early relapse. The long-term goal is to better understand the role of epigenetics in treatment outcomes of MM and develop a non-invasive, clinically convenient approach for predicting relapse. The overall objective of this application is to evaluate 5-hydroxymethylcytosines (5hmC) and 5-methylcytosines (5mC) in circulating cell-free DNA (cfDNA) from MM patients as markers for predicting patients at high risk of early relapse at the time of diagnosis. It is known that greater epigenetic heterogeneity is linked with poorer survival and relapse. We propose that sensitive and robust cfDNA-based epigenetic markers may offer greater convenience and minimal invasiveness for predicting early relapse in MM. In addition to 5mC, changes in 5hmC, an abundant and stable modified cytosine with a distinct gene regulatory function from 5mC, have been implicated in cancer development and pathobiology. However, due to technological constraints, previous studies of cancer epigenetics have largely interpreted modified cytosines as 5mC only, and no study has evaluated the distinct roles of 5hmC and 5mC in the therapeutic significance for MM. To fill the current research and technical gaps and consider the high impact of developing a minimally invasive blood test for MM, we will utilize a highly sensitive and robust technique developed by our team, the nano-Seal-Seq (a chemical labeling technique integrated with the next-generation sequencing), to accurately determine 5hmC and 5mC profiles in cfDNA and CD138+ myeloma ?cancer? cells from bone marrow. The central hypothesis is that the 5hmC/5mC signatures in cfDNA at diagnosis can distinguish MM patients by early relapse status. In Aim 1, we will profile 5hmC and 5mC in cfDNA from ~240 MM patients (relapse within 12 months [early relapse, n=120] vs. >12 months [late relapse, n=120] of starting initial therapy) to develop an integrated predictive index for early relapse. We will validate the 5hmC/5mC markers in two independent replication population of ~750 MM patients. In Aim 2, we will profile 5hmC/5mC in 300 genomic DNA from CD138+ myeloma cells from patients with paired plasma cfDNA to determine a bone marrow-based predictive index, of which the performance will be compared with that of the cfDNA markers. In Aim 3, we will investigate change of 5hmC/5mC in serial cfDNA over time in 130 patients to characterize its therapeutic significance. The proposed research is highly significant, because it is expected to vertically advance understanding of the biological basis for early relapsed MM and promote the development of a new paradigm for epigenetic monitoring (and eventually, treatment) that has broad translational importance in the personalized management of high-risk patients.